Official documentation for Goodeye, an outcome-aligned AI workflow registry. You author workflows as markdown runbooks tagged with the business outcome they serve, pair them with verifiers that score an AI agent against a measurable result, and share them publicly as templates. Everything is reachable from three peer surfaces: the CLI, an MCP server, and a REST API.
Goodeye turns a stated business outcome into a deployed AI workflow. The chain is: Outcome to KPI(s) to Task to Workflow plus Verifiers.
- Workflow: a markdown runbook stored privately in the registry, tagged with the outcome it serves. An agent fetches the body and executes it as a runbook.
- Template: the public form of a workflow, shared under your handle so other people and their agents can find, fetch, and fork it.
- Verifier: a check the workflow runs on agent output. Structural, functional, or semantic (an LLM judge calibrated with examples).
- Image generator: a deployed, owner-scoped image generation capability a workflow can call.
The rendered docs live at goodeye.dev/docs. This repository is the canonical markdown source for that site and for AI coding assistants via Context7. Pages:
| Page | Description |
|---|---|
| Overview | Core concepts and the Goodeye mental model |
| Getting Started | Install the CLI, sign in, fetch and run your first template |
| CLI | Command reference for the goodeye CLI |
| MCP | Connect AI assistants via the Model Context Protocol |
| REST API | Use Goodeye from the /v1 REST API |
| Workflows | Author, version, teach, optimize, and share workflows |
| Templates | Publish, fork, and manage public templates |
| Verifiers | Score agent output with structural, functional, and semantic checks |
| Image Generators | Deploy and run owner-scoped image generation |
| Teams | Collaborate with teams, grants, and invitations |
| Accounts and Billing | Handles, API keys, usage, tiers, and credits |
| Referrals | Invite new users and earn bonus credits on both sides |
- Product: goodeye.dev
- Rendered docs: goodeye.dev/docs
- MCP server URL:
https://mcp.goodeye.dev/mcp - REST API base:
https://api.goodeye.dev/v1 - CLI:
pipx install goodeye(source: Goodeye-Labs/goodeye-cli)
Goodeye is built for AI agents acting on a user's behalf. When an agent fetches a workflow or template body, it follows that body as the user's runbook: it executes the instructions rather than summarizing them.
These docs are indexed by Context7 so AI coding assistants can pull accurate, up-to-date Goodeye usage instructions instead of guessing.
The documentation in this repository is proprietary to Goodeye Labs and provided under the Goodeye Documentation License. You may read and use it to work with Goodeye, allow AI and documentation tools such as Context7 to index it, and freely reuse the code samples. Redistributing or creating derivative documentation is not permitted. See LICENSE for details.